Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by satvikpendem on Thursday March 16, 2023 @08:41PM
(#35191131)
I'm trying the 13B one, it's decent, I'd say on par with GPT 3, definitely not 3.5 or 4 yet however. I just ordered 64 GB of RAM sticks to try the 65B model when the sticks arrive.
[ Parent
| on HN ]
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by dvt on Thursday March 16, 2023 @08:51PM
(#35191230)
How well does it work on DRAM? I thought everyone running heavier models are doing it on GPUs.
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by satvikpendem on Thursday March 16, 2023 @09:12PM
(#35191413)
Alpaca works entirely off CPU so the models should work fine. I'll try tomorrow and report back.
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by BoorishBears on Friday March 17, 2023 @04:28AM
(#35194581)
Recent advances in ML have finally redeemed my decision to building a gaming PC with 64GB of RAM and a 3090 when my most played game was Dolphin Emulator.
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by bdsa on Friday March 17, 2023 @05:16AM
(#35194859)
I forgot briefly that Dolphin refers to GameCube/Wii and was very curious about what a "dolphin emulator" would be (and how it would differ from a dolphin simulator)
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by lbltavares on Saturday March 18, 2023 @08:07PM
(#35214527)
LOL same here
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by Magicianx69 on Monday March 20, 2023 @02:12AM
(#35227887)
man i only play Minecraft an i didn't bother setting it up to use gpu it friggin runs in cpu. so my gpu is not even woken up from slumber yet
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by unshavedyak on Friday March 17, 2023 @10:10AM
(#35197294)
How are you trying it? Alpaca? Or just straight llama.cpp?
Re:Show HN: Alpaca.cpp – Run an Instruction-Tuned Chat-Style...
by satvikpendem on Friday March 17, 2023 @12:04PM
(#35198890)
There's a project called Dalai with a web frontend so you can call npx dalai llama (heh) to get it running. But of course they're all running LLaMA underneath.